(S)-Pregabalin, (S)-(+)-3-(aminomethyl)-5-methylhexanoic acid, a compound having the chemical structure,
is also known as γ-amino butyric acid or (S)-3-isobutyl GABA. (S)-Pregabalin, marketed under the trade name LYRICA®, has been found to activate GAD (L-glutamic acid decarboxylase). (S)-Pregabalin has a dose dependent protective effect on-seizure, and is a CNS-active compound. (S)-Pregabalin is useful in anticonvulsant therapy, due to its activation of GAD, promoting the production of GABA, one of the brain's major inhibitory neurotransmitters, which is released at 30 percent of the brains synapses. (S)-Pregabalin has analgesic, anticonvulsant, and anxiolytic activity.
Several processes for the synthesis of (S)-Pregabalin are known. For example, see DRUGS OF THE FUTURE, 24 (8), 862-870 (1999). One such process is illustrated in scheme 1.

In Scheme 1, 3-isobutyl glutaric acid, compound 2, is converted into the corresponding anhydride, compound 3, by treatment with refluxing acetic anhydride. The reaction of the anhydride with NH4OH produces the glutaric acid mono-amide, compound 4, which is resolved with (R)-1-phenylethylamine, yielding the (R)-phenylethylamine salt of (R)-3-(carbamoylmethyl)-5-methylhexanoic acid, compound 5. Combining the salt with an acid liberates the R enantiomer, compound 6. Finally, a Hoffmann degradation with Br2/NaOH provides (S)-Pregabalin. A disadvantage of this method is that it requires separating the two enantiomers, thereby resulting in the loss of half the product, such that the process cost is high.
Several stereoselective processes for the synthesis of (S)-Pregabalin have been disclosed. For example, U.S. Pat. No. 5,599,973 discloses the preparation of (S)-Pregabalin using stoichiometric (+)-4-methyl-5-phenyl-2-oxazolidinone as a chiral auxiliary that may be recycled. In general, however, that route is of limited use for scale-up, principally due to the low temperature required for the reactions, the use of pyrophoric reagent, such as, butyl lithium, to side reactions, and due to a low overall yield.
Another process is disclosed in U.S. Patent Application Publication No. 2003/0212290, which discloses asymmetric hydrogenation of a cyano-substituted olefin, compound 7, to produce a cyano precursor of (S)-3-(aminomethyl)-5-methyl hexanoic acid, compound 8, as seen in scheme 2.

Subsequent reduction of the nitrile in compound 8 by catalytic hydrogenation produces (S)-Pregabalin. The cyano hexenoate starting material, compound 7, is prepared from 2-methyl propanal and acrylonitrile (Yamamoto et al, Bull. Chem. Soc. Jap., 58, 3397 (1985)). However, the disclosed method requires carbon monoxide under high pressure, raising serious problems in adapting this scheme for production scale processes.
A process published by G. M. Sammis, et al., J. Am. Chem. Soc., 125(15), 4442-43 (2003), takes advantage of the asymmetric catalysis of cyanide conjugate addition reactions. The method discloses the application of aluminum salen catalysts to the conjugate addition of hydrogen cyanide to α,β-unsaturated imides as shown in scheme 3. Reportedly, TMSCN is a useful source of cyanide that can be used in the place of HCN. This process is not practicable for large scale production due to the use of highly poisonous reagents. Moreover, the last reductive step requires high pressure hydrogen, which only adds to the difficulties required for adapting this scheme for a production scale process.

In 1989, Silverman reported a convenient synthesis of 3-alkyl-4-amino acids compounds in SYNTHESIS (1989, 955). Using 2-alkenoic esters as a substrate, a series of GABA analogs were produced by Michael addition of nitromethane to α,β-unsaturated compounds, followed by hydrogenation at atmospheric pressure of the nitro compound to amine moiety as depicted in scheme 4.

Further resolution of compound 14 may be employed to resolve Pregabalin. This, of course, results in the loss of 50 percent of the product.
Recent studies have indicated that cinchona alkaloids are broadly effective in chiral organic chemistry. A range of nitroalkenes were reportedly treated with dimethyl or diethyl malonate in tetrahydrofuran in the presence of cinchona alkaloids to provide high enantiomeric selectivity of compound 15,
and its analogues. For example, see H. Li, et al., J. Am. Chem. Soc., 126(32), 9906-07 (2004). These catalysts are easily accessible from either quinine or quinidine, and are reportedly highly efficient for a synthetically C—C bond forming asymmetric conjugate addition as shown in scheme 5.

R3 represents several alkyl and aryl groups. The scope of the reaction has been extended to other nitroolefins and applied to prepare ABT-546 employing bis(oxazoline)Mg(OTf)2. See, for example, D. M. Barnes, et al., J. Am. Chem. Soc., 124(44), 13097-13105 (2002).
Other groups have investigated a new class of bifunctional catalysts bearing a thiourea moiety and an amino group on a chiral scaffold. See T. Okino, et al., J. Am. Chem. Soc., 127(1), 119-125 (2005). On the basis of a catalytic Michael addition to the nitroolefin with enantiomeric selectivity, they were able to prepare a series of analogues of compound 15.
Thus, there is a need in the art for new processes for the preparation of (S)-Pregabalin that do not suffer from the disadvantages mentioned above.